This issue is rge driver specific, I think. You could try to use gani driver for this and see if it works with HW-checksumming enabled. For this, modify /etc/driver_aliases:
from: rge "pci10ec,8167" rge "pci16ec,116" rge "pci10ec,8136" rge "pciex10ec,8136" to: gani "pci10ec,8167" gani "pci16ec,116" gani "pci10ec,8136" gani "pciex10ec,8136" and reboot. For minimalistic static IP configuration do this: echo "1.1.1.2 netmask 255.255.255.0 mtu 1500 broadcast + up" > /etc/hostname.rge0 echo "1.1.1.1" > /etc/defaultrouter svcadm restart network/physical:default On Fri, 2008-12-26 at 09:11 +0530, Anil Gulecha wrote: > On Fri, Dec 26, 2008 at 7:55 AM, Peter Wright <[email protected]> wrote: > > > > Thanks Tim, Erast, > > > > On 21/12 23:02:24, Erast Benson wrote: > >> Ah... right. Just copy over 2 .deb files: myamanet-gani_*.deb and > >> sunwrge_*.deb and install with dpkg -i *.deb. Should work. > > > > On 22/12 01:52:23, Tim Spriggs wrote: > >> Since there is no working network connectivity to the host you may > >> need to download the files manually. You can grab via the "dl" link > >> provided on these pages: > >> > >> http://builder.tajinc.org/?f=package_status&view_method=name&package_name=myamanet-gani > >> http://builder.tajinc.org/?f=package_status&view_method=name&package_name=sunwrge > > > > This approach did work in one sense - after installing the packages > > and rebooting, I found the network device (rge0) was identified and > > DHCP-configured (now I just need to work out how to configure it as a > > static IP... hm, must be docs somewhere :)). > > > > The problem after that was a little more bizarre - "ping -s $HOST" > > would delay for 10 or so seconds and then spray a group of results on > > the screen at once. Trying to ssh to other machines on my network > > would timeout - running ssh in max-debug mode (ie. 'ssh -v -v -v') > > indicated that it got up to a certain point and then no further. > > > > I could even see output in the logs on the other machines (that I was > > trying to ssh to) showing an initial connection from the Nexenta > > machine, but nothing beyond that. > > > > And similarly, apt-get/aptitude update would just timeout. > > > > I eventually guessworked my way to thinking that it might be the same > > problem described in this OpenSolaris bug: > > > > http://bugs.opensolaris.org/view_bug.do?bug_id=6717107 > > > > This suggested a workaround as follow: > > > >> Add > >> set ip:dohwcksum = 0 > >> to /etc/system. > >> > >> Or run this command as root: > >> echo 'ip`dohwcksum/w 0' | mdb -kw > > > > ...which worked. > > > > Well, worked in the sense that I now appear to have a functioning > > network :-). > > > > I gathered from some of the places referencing this bug that disabling > > hardware-checksumming is probably not a good long-term solution, but > > I'm quite happy to settle for a short-term solution at the moment. > > > > The bug report indicates it was fixed in snv_99, which I presume is > > not part of the current Nexenta2 beta release (but hopefully will be at > > some point in the future :)). > > > > Hmm.. the beat release is b104+, so the issue has either shown up > again, or (sightly) different. _______________________________________________ gnusol-users mailing list [email protected] http://lists.sonic.net/mailman/listinfo/gnusol-users
